From: Richard Biener <richard.guenther@gmail.com>
To: "Stubbs, Andrew" <andrew_stubbs@mentor.com>
Cc: GCC Development <gcc@gcc.gnu.org>
Subject: Re: DSE and maskstore trouble
Date: Tue, 03 Jul 2018 13:52:00 -0000 [thread overview]
Message-ID: <CAFiYyc2V6bpP4O+xG7KZvtrf2Jb_k7uqyp+xtzq__iwTfaEqBQ@mail.gmail.com> (raw)
In-Reply-To: <bfcc6556-7f08-069c-7165-8706e9f2865b@mentor.com>
On Tue, Jul 3, 2018 at 2:46 PM Andrew Stubbs <andrew_stubbs@mentor.com> wrote:
>
> On 03/07/18 13:21, Richard Biener wrote:
> > Ok, so if we vectorize the above with 64 element masked stores
> > then indeed the RTL representation is _not_ safe. That is because
> > while the uses in the masked stores should prevent things from
> > going bad there is also TBAA to consider which means those
> > uses might not actually _be_ uses (TBAA-wise) of the earlier
> > stores. In the above case rtattr * doesn't alias int (or whatever
> > types rta_type or rta_len have). That means to DSE the earlier
> > stores are dead.
>
> I managed to get it to generate maskstore without the unspec, and the
> code now runs correctly.
OK, that is good.
> I don't follow your AA reasoning. You say the use stops it being bad,
> and then you say the stores are dead, which sounds bad, yet it's not
> deleting them now.
If you look at RTL dumps (with -fstrict-aliasing, thus -O2+) you should
see MEM_ALIAS_SETs differing for the earlier stores and the masked
store uses.
Now I'm of course assuming DSE is perfect, maybe it isn't ... ;)
Richard.
>
> Confused. :-(
>
> Andrew
next prev parent reply other threads:[~2018-07-03 13:52 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-03 9:59 Andrew Stubbs
2018-07-03 10:16 ` Richard Biener
2018-07-03 10:33 ` Andrew Stubbs
2018-07-03 10:57 ` Andrew Stubbs
2018-07-03 11:02 ` Richard Biener
2018-07-03 11:06 ` Andrew Stubbs
2018-07-03 11:31 ` Richard Biener
2018-07-03 11:38 ` Andrew Stubbs
2018-07-03 11:46 ` Richard Biener
2018-07-03 11:57 ` Andrew Stubbs
2018-07-03 12:21 ` Richard Biener
2018-07-03 12:46 ` Andrew Stubbs
2018-07-03 13:52 ` Richard Biener [this message]
2018-07-03 15:19 ` Andrew Stubbs
2018-07-03 16:45 ` Richard Biener
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAFiYyc2V6bpP4O+xG7KZvtrf2Jb_k7uqyp+xtzq__iwTfaEqBQ@mail.gmail.com \
--to=richard.guenther@gmail.com \
--cc=andrew_stubbs@mentor.com \
--cc=gcc@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).